|
on Discrete Choice Models |
| By: | James J. Heckman; John Eric Humphries; Gregory Veramendi; Sergio Urzúa |
| Abstract: | This paper develops and estimates a model with multiple schooling choices that identifies the causal effect of different levels of schooling on health, health-related behaviors, and labor market outcomes. We develop an approach that is a halfway house between a reduced form treatment effect model and a fully formulated dynamic discrete choice model. It is computationally tractable and identifies the causal effects of educational choices at different margins. We estimate distributions of responses to education and find evidence for substantial heterogeneity in unobserved variables on which agents make choices. The estimated treatment effects of education are decomposed into the direct benefits of attaining a given level of schooling and indirect benefits from the option to continue on to further schooling. Continuation values are an important component of our estimated treatment effects. While the estimated causal effects of education are substantial for most outcomes, we also estimate a quantitatively important effect of unobservables on outcomes. Both cognitive and socioemotional factors contribute to shaping educational choices and labor market and health outcomes. We improve on LATE by identifying the groups affected by variations in the instruments. We find benefits of cognition on most outcomes apart from its effect on schooling attainment. The benefits of socioemotional skills on outcomes beyond their effects on schooling attainment are less precisely estimated. |
| JEL: | C32 C38 I12 I14 I21 |
| Date: | 2014–03 |
| URL: | http://d.repec.org/n?u=RePEc:nbr:nberwo:19971&r=dcm |
| By: | McCAUSLAND, William; MARLEY, A. A. J. |
| Abstract: | We complete the development of a testing ground for axioms of discrete stochastic choice. Our contribution here is to develop new posterior simulation methods for Bayesian inference, suitable for a class of prior distributions introduced by McCausland and Marley (2013). These prior distributions are joint distributions over various choice distributions over choice sets of different sizes. Since choice distributions over different choice sets can be mutually dependent, previous methods relying on conjugate prior distributions do not apply. We demonstrate by analyzing data from a previously reported experiment and report evidence for and against various axioms. |
| Keywords: | Random utility, discrete choice, Bayesian inference, MCMC |
| JEL: | C11 C35 C53 D01 |
| Date: | 2013 |
| URL: | http://d.repec.org/n?u=RePEc:mtl:montde:2013-06&r=dcm |
| By: | Palma, David; Dios Ortuzar, Juan de; Casaubon, Gerard; Rizzi, Luis I.; Agosin, Eduardo |
| Abstract: | Wine is a complex product. Preferences for it are not only highly heterogeneous throughout the population, but also amply susceptible to context. The objective of this study is to discover and measure these preferences, focusing on a set of non-sensory attributes of wine. To identify the most relevant non-sensory attributes of wine, from the consumers’ standpoint we considered four sources: existing literature, a Delphi survey (applied to wine marketing experts), in- depth interviews and a web-page survey answered by fairly large sample of wine consumers. Not all sources were consistent on which attributes were the most important. Notably, consumers did not select price as a relevant attribute on the web survey, even though it had been considered relevant in the in-depth interviews. Finally, six wine attributes were selected for inclusion in a stated choice (SC) experiment: grape variety, alcohol level, label design, product recommendations, price and discounts. An efficient experimental design was then developed and a web based SC survey was applied to 274 regular wine consumers (who had already answered the previous web survey). These consumers have high income (among the richest 20% of the Chilean population), only 28% of them are female and 33% are 35 years old or younger. The SC experiment simulated a purchase, at a retail store, for a casual meal with friends. A fixed fictional brand was used for all the wines presented on the experiment. With this data we estimated various discrete choice models, including mixed logit and hybrid choice models. Grape variety was found to be the main driver of preferences. Evidence of preference for higher alcohol levels was also discovered. Price proved to be highly endogenous, as it is strongly related to wine’s expected quality. Recommendation by a friend and critics were equally valued, except in the case of (self-declared) expert consumers. Preferences over label designs showed high variability. The results also suggest that price is a key attribute in the construction of expected quality by the consumer before tasting the product. |
| Keywords: | Hybrid Discrete Choice Models, Consumer choice, wine, Consumer/Household Economics, Demand and Price Analysis, |
| Date: | 2013–07 |
| URL: | http://d.repec.org/n?u=RePEc:ags:aawewp:164855&r=dcm |
| By: | Bertrand Candelon; Christophe Hurlin; Elena Dumitnescu |
| Abstract: | Traditionally, nancial crisis Early Warning Systems (EWSs) rely on macroeconomic leading indicators to forecast the occurrence of such events. This paper extends such discrete-choice EWSs by taking into account the persistence of the crisis phenomenon. The dynamic logit EWS is estimated using an exact maximum likelihood estimation method both in a time series and panel form. This model's forecasting abilities are then scrutinized by using an evaluation methodology recently designed specically for EWSs. When applied for predict- ing currency crises for 16 countries, this new EWS turns out to exhibit signicantly better predictive abilities than the existing static one, both in- and out-of -sample, thus supporting the use of dynamic specications for EWSs for nancial crises. |
| Keywords: | dynamic models, currency crisis, Early Warning System. |
| JEL: | C33 F37 |
| Date: | 2014–02–25 |
| URL: | http://d.repec.org/n?u=RePEc:ipg:wpaper:2014-161&r=dcm |
| By: | Marc K Chan (Economics Discipline Group, University of Technology, Sydney) |
| Abstract: | We use data from Florida Transition Program, a welfare reform experiment in the 1990s, to estimate a discrete choice dynamic programming model of labor supply and welfare participation with potentially time-inconsistent individuals. The time preference parameters are identified through exclusion restrictions generated by welfare time limits. Around one-fourth of the individuals can be regarded as present-biased, and they exhibit a low degree of naivety. Time-inconsistency reduces income by 15 percent and the net tax contribution by almost half. Present-biased individuals are generally more responsive to policy changes than time-consistent individuals. By aggravating the commitment problem, an increase in welfare benefits reduces utility from a time-consistent perspective. An expansion of Earned Income Tax Credit (EITC) can be revenue-neutral due to cross-subsidization between present-biased and time-consistent individuals. A “prowork time limit” is proposed as a more incentivizing policy than standard time limits. A dynamic nonwork tax that is triggered by past employment can generate strong commitment-related incentives and increase utility from a time-consistent perspective. The nonwork tax can be implemented as a targeting intervention, as an estimated 70 percent of present-biased individuals will adopt the policy as a commitment device. |
| Keywords: | Welfare dependence; hyperbolic discounting; time limits; female labor supply; welfare reform; policy experiment; discrete choice dynamic programming |
| JEL: | I3 C3 J2 |
| Date: | 2014–03–01 |
| URL: | http://d.repec.org/n?u=RePEc:uts:ecowps:19&r=dcm |
| By: | Santosh Kumar (Department of Economics and International Business, Sam Houston State University); Emily Dansereau (University of Washington); Chris Murray (University of Washington) |
| Abstract: | In this paper, we examine if access to health facilities improves institutional birth delivery in a resource-constrained country like India. Using a household- and village-level health survey, we find that women living closer to health facilities have a higher probability of in-facility births. A one kilometer increase in the distance to the nearest health facility decreases the probability of institutional delivery by 0.8%. This result does not change significantly even after we account for endogenous placement of health facilities. The results of Two-Stage Residual Inclusion (2SRI) and IV-Probit models suggest that an additional travel of one kilometer decreases probability of in-facility delivery (IFD) by 4.4%. The policy simulation result suggest that, the mean probability of in-facility delivery increases when the density of health facility is increased. Overall, results suggest that geographic distance is an important barrier to service utilization and improving access to health facilities may be an important policy instrument to improve utilization of health services in resource-poor countries. |
| Date: | 2014–03 |
| URL: | http://d.repec.org/n?u=RePEc:shs:wpaper:1405&r=dcm |
| By: | Bernd Hayo; Florian Neumeier |
| Abstract: | The poor state of public finances in many countries has led to calls for fiscal consolidation. In practice, implementing concrete consolidation measures appears to meet with public resistance, suggesting that the success of consolidation efforts strongly depends on the popularity of the chosen measures. To identify public attitudes toward fiscal consolidation and alternative consolidation measures, we conducted a survey among 2,000 German citizens. Applying ordered and multinominal logit models, we test theory-based hypotheses about the determinants of individual attitudes toward public debt. We find that, inter alia, personal economic situation, time preferences, fiscal illusion, and trust in politicians exert a significant impact on attitudes toward fiscal consolidation and preferences for alternative consolidation measures. |
| Keywords: | Public debt; sovereign debt crisis; public attitude; Germany; fiscal consolidation |
| JEL: | D72 H31 H63 |
| Date: | 2014–03–19 |
| URL: | http://d.repec.org/n?u=RePEc:sol:wpaper:2013/158953&r=dcm |
| By: | Bernd Hayo; Matthias Neuenkirch |
| Abstract: | We study how financial market participants process news from four major central banks—the Bank of England (BoE), the Bank of Japan (BoJ), the European Central Bank (ECB), and the Federal Reserve (Fed), using a novel survey of 450 financial market participants from around the world. Our results indicate that, first, respondents rely more on newswire services to learn about central bank events than on self-monitoring. In general, the Fed is watched most closely, followed by the ECB, the BoE, and the BoJ. Second, we estimate ordered probit models to relate the two different types of central bank watching to the perceived importance of central bank events and the reliability of media coverage. Our results indicate that financial agents have to rely on newswire services to appropriately cope with a globalised market environment and digest news. However, when respondents consider an event particularly important, they tend to self-monitor it, especially when the event is taking place in their home region. |
| Keywords: | Central Bank Communication, Financial Market Participants, Information Processing, Interest Rate Decisions, Newswire Services, Reliability, Survey |
| JEL: | D83 E52 E58 |
| Date: | 2014 |
| URL: | http://d.repec.org/n?u=RePEc:trr:wpaper:201407&r=dcm |
| By: | Daniel Montolio (Universitat de Barcelona & IEB); Simón Planells-Struse (Universitat de Barcelona & IEB) |
| Abstract: | Crime risk perception is known to be an important determinant of individual well-being. It is therefore crucial that we understand the factors affecting this perception so that governments can identify the (public) policies that might reduce it. Among such policies, public resources devoted to policing emerge as a key instrument not only for tackling criminal activity but also for impacting on citizens’ crime risk perception. In this framework, the aim of this study is to analyze both the individual and neighbourhood determinants of citizens’ crime risk perception in the City of Barcelona (Spain) focusing on the effect of police proximity and taking into account the spatial aspects of neighbourhood characteristics. After controlling for the possible problems of the endogeneity of police forces and crime risk perception and the potential sorting of individuals across neighbourhoods, the results indicate that crime risk perception is reduced when non-victims exogenously interact with police forces. Moreover, neighbourhood variables, such as proxies of social capital and the level of incivilities, together with individual characteristics have an impact on citizens’ crime risk perception. |
| Keywords: | Crime risk perception, police forces, multilevel ordered logit model |
| JEL: | C21 H50 K42 |
| Date: | 2014 |
| URL: | http://d.repec.org/n?u=RePEc:ieb:wpaper:2013/6/doc2014-1&r=dcm |
| By: | Jérôme Danguy |
| Abstract: | Using patent-based indicators, this paper aims at explaining to what extent the productionof innovation is globalized. Firstly, it provides evidence – over time, across countriesand across industrial sectors – on the patterns in international technological collaborationand in cross-border ownership of innovation. Secondly, a fractional logit model is estimatedfor a unique panel dataset covering patent information of 21 industries in 29 countries from1980 to 2005. The results show that countries tend to be more globalized in industrial sectorsin which they are less technologically specialized. It suggests that globalization of innovationis more driven by home-base augmenting determinants than home-base exploitingones. The empirical findings also indicate that the intensity of globalization of innovationis higher in multidisciplinary country-industry pairs and in those which compete internationallyin trade. |
| Keywords: | internationalization, R&D collaboration, patent statistics, industrial sectors |
| Date: | 2014–02–23 |
| URL: | http://d.repec.org/n?u=RePEc:ict:wpaper:2013/157283&r=dcm |
| By: | Mundra, Kusum (Rutgers University); Sharma, Amarendra (Elmira College) |
| Abstract: | Home adequacy for different groups in the U.S. has not been adequately studied. Using the data from the national level American Housing Survey for the year 2009and logit model, this paper finds that there is a significant adequacy difference for Blacks and Hispanics when compared to whites in the U.S. However, that is not the case for immigrants relative to the natives. We also find that then naturalization improves housing adequacy among immigrant homeowners, whereas, the female headed households have a significantly higher home adequacy than that of the male headed households. Similar to the homeownership findings, this paper highlights that the public policies should aim to narrow the home adequacy gap between whites and minorities and encourage naturalization to improve adequacy among immigrant homeowners. |
| Keywords: | housing adequacy gap structural adequacy, U.S. residential real estate, immigrants, minorities, American Housing Survey, naturalized |
| JEL: | R2 J15 |
| Date: | 2014–03 |
| URL: | http://d.repec.org/n?u=RePEc:iza:izadps:dp8038&r=dcm |
| By: | María Teresa Costa (Universitat de Barcelona & IEB); José García-Quevedo (Universitat de Barcelona & IEB); Agustí Segarra (GRIT, CREIP, Rovira i Virgili University) |
| Abstract: | This paper examines the extent to which innovative Spanish firms pursue improvements in energy efficiency (EE) within their innovation objectives. The increase in energy consumption and its impact on greenhouse gas emissions justifies the greater attention being paid to energy efficiency and especially to industrial EE. The ability of manufacturing companies to innovate and improve their EE has a substantial influence on reaching the objectives regarding climate change mitigation. Despite the effort to design more efficient energy policies, the EE determinants in manufacturing firms have been little studied in the empirical literature. From an exhaustive sample of Spanish manufacturing firms and using a probit model, we examine the energy efficiency determinants to those firms that have innovated. To carry out the econometric analysis, we use a panel data coming from CIS (Community Innovation Survey) for the period 2008-2011 that includes 4,458 manufacturing firms. Among firm characteristics, the empirical results underline the importance of size in facilitating the adoption of technology that improves energy efficiency; while among the factors related to companies’ behavior, the favorable influence of organizational innovations and innovations related with the reduction of environmental impacts stand out as the main factors in carrying out innovations with the objective of increasing energy efficiency. |
| Keywords: | Energy efficiency, corporate targets, innovation, Community Innovation Survey |
| JEL: | Q40 Q55 O31 |
| Date: | 2014 |
| URL: | http://d.repec.org/n?u=RePEc:ieb:wpaper:2013/6/doc2014-10&r=dcm |
| By: | Goriunov Dmytro; Venzhyk Katerina |
| Abstract: | Using a large proprietary dataset provided by the tenth largest Ukrainian banking institution, we posit reasons for loan defaults within two major groups of retail borrowers; car loans and mortgages. Two model types were used, namely logistic regression and neural networks. The results of our estimations suggest that a) data currently collected by banks are sufficient to predict defaults, but bankers should collect more information, and that b) the neural networks model slightly outperforms the logit model in predictive power. |
| JEL: | G21 G32 G33 |
| Date: | 2013–05–23 |
| URL: | http://d.repec.org/n?u=RePEc:eer:wpalle:13/07e&r=dcm |
| By: | Diego Dueñas Fernández; Carlos Iglesias Fernández; Raquel Llorente Heras |
| Abstract: | El presente artículo tiene como objetivo determinar cuál es la desigual influencia que tienen en el mercado de trabajo español de los nuevos aspectos pre-market: factores sociales y sociológicos que influyen en la propia personalidad y que condicionan de manera diferente las decisiones laborales de hombres y mujeres. Concretamente nuestro objeto de estudio es la segregación laboral y las diferencias salariales por género. Para ello, el trabajo analiza los datos de la Encuesta de Clases Sociales y Estructura Social elaborada por el Centro de Investigaciones Sociológicas (CIS) en 2006. Con respecto a la segregación ocupacional, los resultados alcanzados sobre la estimación de una serie de probits multinivel indican que la segregación ocupacional femenina se fundamenta en las decisiones personales y el tipo de ubicación laboral detentada más que en los aspectos pre-market que no consiguen alcanzar una influencia significativa. Los resultados alcanzados en las estimaciones salariales sobre la descomposición de Oaxaca-Blinder establecen que los aspectos relacionados con el puesto de trabajo siguen siendo los factores determinantes de la desigual remuneración por género existente en España. Asimismo, los aspectos pre-market presentan una reducida aportación a la explicación de las diferencias salariales. Todo ello, nos lleva finalmente a pensar que los aspectos pre-market no son el determinante directo de la segregación ocupacional y del gap salarial en España sino que contribuyen levemente a su explicación. |
| Keywords: | aspectos pre-market, segregación laboral, diferencias salariales por género. |
| Date: | 2014–03 |
| URL: | http://d.repec.org/n?u=RePEc:uae:wpaper:0414&r=dcm |